MLA-C01 Question 191
Select 3You are training a Random Forest model to predict customer churn for a telecommunications company. The model is underfitting the training data, leading to poor performance on both the training and validation datasets. Which hyperparameter adjustments are most likely to address this issue?
- A
Increase the number of trees in the Random Forest model
- B
Decrease the maximum depth of each tree
- C
Increase the maximum depth of each tree
- D
Reduce the minimum number of samples required to split a node
- E
Increase the minimum number of samples required to split a node
Show answer and explanation
Correct answers: A, C, D
Explanation
Underfitting occurs when a model is too simple to capture the underlying patterns in the data. In the case of a Random Forest, increasing the number of trees helps reduce variance, while increasing the maximum depth of each tree and reducing the minimum number of samples required to split a node allow the model to learn more complex patterns. These adjustments collectively help mitigate the underfitting issue and improve the model's performance.
- A. Correct.
Increasing the number of trees in the Random Forest model can improve its performance by reducing variance and allowing the model to capture more patterns in the data.
- B. Incorrect.
Decreasing the maximum depth of each tree would likely worsen the underfitting issue since it further restricts the model's ability to learn complex patterns.
- C. Correct.
Increasing the maximum depth of each tree allows the model to capture more complex patterns in the data, which can help address underfitting.
- D. Correct.
Reducing the minimum number of samples required to split a node enables the model to create more splits and learn finer patterns in the data, helping to mitigate underfitting.
- E. Incorrect.
Increasing the minimum number of samples required to split a node would likely worsen the underfitting issue as it restricts the model's ability to grow and capture finer details.